In this video, we're going to take some time to think broadly about the methods or the approaches that are available to us to change behavior and organization. And we're going to think broadly, we're going to think about systems, and we're going to think about structure processes for changing behavior. One approach to consider is the administrative approach to change. This involves mostly top down mandates implemented through policy, relying on a very tight hierarchy. This type of change is not participatory from staff at lower levels of the organization, does not provide much sense of autonomy or control over the work from frontline staff. It does not easily accommodate differences in work areas or customization to different work context. So this is a common approach. It's an old approach and unfortunately it does not, on its own, meet many of the needs we have for patients' safety and quality improvement efforts. So it's something we want to contrast a different approach to, something that is sensitive to a broader range of factors and addresses those broader range of factors. And again we want something that is participatory and that does pull people together in a sense of improvement and accomplishment. So we can think about a work system having multiple layers or multiple components. And in each of those components of the system, there are some factors that are going to influence whether a change happens or not. So at the individual level, this can be the patient, this can be staff, healthcare providers, administrative staff. There's a knowledge and the skills and the attitudes that they have. So are they capable of the new performance expectations? Do they think that those new performance expectations are appropriate or important? Those will determine whether you're are effective or not. The actual tasks involved, the workload, the complexity and again the autonomy that people may have, if we're implementing a new change where people are already overloaded in the tasks that they have that will potentially create some problems for us. The technology, the tools that people use, is it easy to use? Does it fit the workflow that people have? These things are going to influence adoption and we see that very commonly in lots of different health I.T. systems. The environment that this change is supposed to happen in, are there distractions? Is it noisy? What's the physical space look like? Does it make it more or less effortful for people to perform these new expectations? Those will impact us. And of course, the organization which we talked about a little bit already with our administrative approach. So policies, accountability, leadership and the organizational culture can all influence whether we're successful or not. And we need to think about each of these works system components as we design our interventions to make change happen. So the factors at each level of the system have different methods to address them. For example, at the individual level, we have training programs to build skill and knowledge and to change attitudes. We have audit and feedback processes that can provide developmental feedback to people and coaching to improve their performance over time. We can use participatory design processes, like we've been talking about throughout this lesson to engage people in making the change for themselves that can change without a doubt people's attitudes about the nature of that change. We have shadowing processes, to learn from other peers about how to do new tasks. The bucket of tasks in the system. The whole domain of lean management strategies can be helpful here to analyze work processes to reduce variation to increase reliability. There's a very mature set of approaches there for managing the task component of the system. Similarly for technology and tools, we can use user-centered design and evaluation practices to get real and user feedback on our designs, on our technology, on the tools we've chosen to use and make sure that we're addressing their needs as effectively as possible. There's the environment, the larger physical space and everything else that's going on within any one given location. We can use in-situ simulations either high fidelity or low fidelity to walk through new process changes and elicit ideas about what we can do to mitigate any risks and to strengthen our approach for managing a change. And of course at the organization level, senior leadership engagement in participation in an effort in the degree to which a change effort aligns with the broader organizational goals and initiatives can all influence whether we're going to be effective or not. So thinking across all of those different layers of the system, what can we do? What can we capitalize on to make sense of those to design the best intervention that we can, and we'll go through a few examples. First, we'll discuss a few principles for designing a new work practice. So these are ideas we can keep in mind as we're thinking through, what is our intervention going to look like, and what can we do to support the end user or the people whose behavior needs to change? What can we do to support them as much as possible and make this change as easy as possible for them? First, use affordances. Affordances make the desired behavior intuitive. As much as possible, we want our equipments, our environment to tell us what is expected of us. This is not always practical but when it is, it's very powerful. Provide just in time support. So we want to make sure people have access to training and learning and instruction while they're doing the actual task in the work environment. We're not relying on what people learned in a training session but their support is there for them wherever they're doing the work. We want to use nudges, we want to make the desired behavior easy and the undesired behavior difficult or impossible to do. Next, provide smart defaults. The right answer should be the first answer people are presented with, and they should have to go out of their way to choose something inappropriate or incorrect. Next, we want to provide feedback. So people need to know how they're performing relative to expectations. This can come in many different forms, but the key piece that people need to understand how their behavior is meeting expectations or not. Next we want to try to minimize cognitive effort. So this is eliminating unnecessary mental tasks, calculations doing math or the memory burden people have. These of course introduce errors but they also make it more work intensive and less likely that people perform a behavior as desired. And the last piece is minimizing physical effort. So, reducing any unnecessary movements or other physical burdens to completing the desired behavior. So those are general principles to think about as we go through our examples. The first example we'll talk about is central line placement. So at the individual level, we may have lapses in performance. This could involve cross contamination as we're setting up a site to insert a central line. Perhaps we touch our nose, perhaps we touch our face and contaminate our gloves. We want to be able to provide feedback in the case that that happens. And that's why we would include an observer in the work process to catch and correct these deviations from protocols that someone doing the task might not notice themselves doing. There's a set of task related factors for inserting a central line. We have maybe evidence based guidelines are not widely disseminated. So there's a lack of clarity of what the task should look like. Some principles we can use are to reduce the cognitive effort involved in understanding that task and to provide just in time support, and a checkslist of the procedural steps can get us there. So that requires people to remember less of what the actual protocol is. And it provides that information while they're in the place doing the task. And finally, the technology and the tools. So, a common challenge in days past for Central Line placement, was that we don't have all the supplies we need or that the supplies are located many different places and sometimes we use the wrong supplies. So, some strategies or principles to use address this involve reducing the physical effort and using smart defaults. And we can do that by creating bundles or line carts or gathers all of our physical materials in one place, so that reduces the physical work we need to do to prepare for the task. If we only include the appropriate supplies, that's a smart default. Right? So the correct solutions are correct supplies and materials are there for us. We have to exert extra effort to use something incorrectly. Another example, we've all seen hygiene. At this level, there's individual factors around the attitudes towards this, and that it might not be a priority for them. People understand hand hygiene as important but there are tradeoffs throughout the day and maybe this doesn't warrant their attention on a regular basis. So there is a way to provide feedback on this. So there's training approaches that make the transmission visible. When we're out training around hand hygiene we can actually see how much gets spread via contact, and we can also provide compliance and infection data to people so that we can make the consequences for hand hygiene compliance or lack there of more real to people. We can focus on tasks too around hand hygiene. As I said out of course, there are many priorities at the bedside, hand hygiene. It takes time, it takes effort. We can do things to reduce the physical effort involved, and that's probably been one of the most successful strategies for hand hygiene compliance by making hand sanitizers at critical points, easily and readily available. So coming in a room, out of the room, at the bedside, at nursing stations making them readily available, so we don't have to spend time finding a sink or finding a small number of dispensers on a unit. And the last piece around hygiene, there are some organizational issues that can be addressed. So, there may be no consequences for having poor hand hygiene compliance. So there are policy, there is a role for policy and leadership in this to create and enforce a policy of corrective action when we find someone who is not adherent. But then also punitive action for when there is chronic recurrent non-compliance. So, in summary, a top down administrative approach to change is limited. There is a role for that as we saw. But there are many other factors that drive change. If we want to be successful, we need to consider that broader range of system factors that are going to influence behavior. And we need to design solutions that are principled, and that address the multitude of systems factors that can impact behavior. And to do that, we need to engage our stakeholders and user insights and wisdom to make sure our interventions are fit for use.
